Development of a Novel In Silico Classification Model to Assess Reactive Metabolite Formation in the Cysteine Trapping Assay and Investigation of Important Substructures DOI Creative Commons
Yuki Umemori, Koichi Handa,

Saki Yoshimura

et al.

Biomolecules, Journal Year: 2024, Volume and Issue: 14(5), P. 535 - 535

Published: April 30, 2024

Predicting whether a compound can cause drug-induced liver injury (DILI) is difficult due to the complexity of drug mechanism. The cysteine trapping assay method for detecting reactive metabolites that bind microsomes covalently. However, it cumbersome use 35S isotope-labeled this assay. Therefore, we constructed an in silico classification model predicting positive/negative outcome We collected 475 compounds (436 in-house and 39 publicly available drugs) based on experimental data performed study, composition results showed 248 positives 227 negatives. Using Message Passing Neural Network (MPNN) Random Forest (RF) with extended connectivity fingerprint (ECFP) 4, built machine learning models predict covalent binding risk compounds. In time-split dataset, AUC-ROC MPNN RF were 0.625 0.559 hold-out test, restrictively. This result suggests has higher predictivity than dataset. Hence, conclude better predictive power. Furthermore, most substructures contributed positively consistent previous results.

Language: Английский

Chemprop: A Machine Learning Package for Chemical Property Prediction DOI Creative Commons
Esther Heid, Kevin P. Greenman, Yunsie Chung

et al.

Journal of Chemical Information and Modeling, Journal Year: 2023, Volume and Issue: 64(1), P. 9 - 17

Published: Dec. 26, 2023

Deep learning has become a powerful and frequently employed tool for the prediction of molecular properties, thus creating need open-source versatile software solutions that can be operated by nonexperts. Among current approaches, directed message-passing neural networks (D-MPNNs) have proven to perform well on variety property tasks. The package Chemprop implements D-MPNN architecture offers simple, easy, fast access machine-learned properties. Compared its initial version, we present multitude new functionalities such as support multimolecule reactions, atom/bond-level spectra. Further, incorporate various uncertainty quantification calibration methods along with related metrics pretraining transfer workflows, improved hyperparameter optimization, other customization options concerning loss functions or atom/bond features. We benchmark models trained using reaction, atom-level, spectra functionality data sets, including MoleculeNet SAMPL, observe state-of-the-art performance water-octanol partition coefficients, reaction barrier heights, atomic partial charges, absorption enables out-of-the-box training problem settings in fast, user-friendly, software.

Language: Английский

Citations

163
Structure-based drug design with geometric deep learning DOI Creative Commons
Clemens Isert, Kenneth Atz, Gisbert Schneider

et al.

Current Opinion in Structural Biology, Journal Year: 2023, Volume and Issue: 79, P. 102548 - 102548

Published: Feb. 25, 2023

Structure-based drug design uses three-dimensional geometric information of macromolecules, such as proteins or nucleic acids, to identify suitable ligands. Geometric deep learning, an emerging concept neural-network-based machine has been applied macromolecular structures. This review provides overview the recent applications learning in bioorganic and medicinal chemistry, highlighting its potential for structure-based discovery design. Emphasis is placed on molecular property prediction, ligand binding site pose de novo The current challenges opportunities are highlighted, a forecast future presented.

Language: Английский

Citations

98
Enabling late-stage drug diversification by high-throughput experimentation with geometric deep learning DOI Creative Commons
David F. Nippa, Kenneth Atz,

Remo Hohler

et al.

Nature Chemistry, Journal Year: 2023, Volume and Issue: 16(2), P. 239 - 248

Published: Nov. 23, 2023

Abstract Late-stage functionalization is an economical approach to optimize the properties of drug candidates. However, chemical complexity molecules often makes late-stage diversification challenging. To address this problem, a platform based on geometric deep learning and high-throughput reaction screening was developed. Considering borylation as critical step in functionalization, computational model predicted yields for diverse conditions with mean absolute error margin 4–5%, while reactivity novel reactions known unknown substrates classified balanced accuracy 92% 67%, respectively. The regioselectivity major products accurately captured classifier F -score 67%. When applied 23 commercial molecules, successfully identified numerous opportunities structural diversification. influence steric electronic information performance quantified, comprehensive simple user-friendly format introduced that proved be key enabler seamlessly integrating experimentation functionalization.

Language: Английский

Citations

42
Prospective de novo drug design with deep interactome learning DOI Creative Commons
Kenneth Atz,

Leandro Cotos,

Clemens Isert

et al.

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: April 22, 2024

Abstract De novo drug design aims to generate molecules from scratch that possess specific chemical and pharmacological properties. We present a computational approach utilizing interactome-based deep learning for ligand- structure-based generation of drug-like molecules. This method capitalizes on the unique strengths both graph neural networks language models, offering an alternative need application-specific reinforcement, transfer, or few-shot learning. It enables “zero-shot" construction compound libraries tailored bioactivity, synthesizability, structural novelty. In order proactively evaluate interactome framework protein design, potential new ligands targeting binding site human peroxisome proliferator-activated receptor (PPAR) subtype gamma are generated. The top-ranking designs chemically synthesized computationally, biophysically, biochemically characterized. Potent PPAR partial agonists identified, demonstrating favorable activity desired selectivity profiles nuclear receptors off-target interactions. Crystal structure determination ligand-receptor complex confirms anticipated mode. successful outcome positively advocates de application in bioorganic medicinal chemistry, enabling creation innovative bioactive

Language: Английский

Citations

32
Open-Source Machine Learning in Computational Chemistry DOI Creative Commons
Alexander Hagg, Karl N. Kirschner

Journal of Chemical Information and Modeling, Journal Year: 2023, Volume and Issue: 63(15), P. 4505 - 4532

Published: July 19, 2023

The field of computational chemistry has seen a significant increase in the integration machine learning concepts and algorithms. In this Perspective, we surveyed 179 open-source software projects, with corresponding peer-reviewed papers published within last 5 years, to better understand topics being investigated by approaches. For each project, provide short description, link code, accompanying license type, whether training data resulting models are made publicly available. Based on those deposited GitHub repositories, most popular employed Python libraries identified. We hope that survey will serve as resource learn about or specific architectures thereof identifying accessible codes topic basis. To end, also include for generating fundamental learning. our observations considering three pillars collaborative work, open data, source (code), models, some suggestions community.

Language: Английский

Citations

28
Exploring protein–ligand binding affinity prediction with electron density-based geometric deep learning DOI Creative Commons
Clemens Isert, Kenneth Atz, Sereina Riniker

et al.

RSC Advances, Journal Year: 2024, Volume and Issue: 14(7), P. 4492 - 4502

Published: Jan. 1, 2024

A deep learning approach centered on electron density is suggested for predicting the binding affility between proteins and ligands. The thoroughly assessed using various pertinent benchmarks.

Language: Английский

Citations

12
Quantum mechanical-based strategies in drug discovery: Finding the pace to new challenges in drug design DOI Creative Commons
Tiziana Ginex, Javier Vázquez,

Carolina Estarellas

et al.

Current Opinion in Structural Biology, Journal Year: 2024, Volume and Issue: 87, P. 102870 - 102870

Published: June 24, 2024

The expansion of the chemical space to tangible libraries containing billions synthesizable molecules opens exciting opportunities for drug discovery, but also challenges power computer-aided design prioritize best candidates. This directly hits quantum mechanics (QM) methods, which provide chemically accurate properties, subject small-sized systems. Preserving accuracy while optimizing computational cost is at heart many efforts develop high-quality, efficient QM-based strategies, reflected in refined algorithms and approaches. QM-tailored physics-based force fields coupling QM with machine learning, conjunction computing performance supercomputing resources, will enhance ability use these methods discovery. challenge formidable, we undoubtedly see impressive advances that define a new era.

Language: Английский

Citations

11
Making the Case for Quantum Mechanics in Predictive Toxicology─Nearly 100 Years Too Late? DOI
Jakub Kostal

Chemical Research in Toxicology, Journal Year: 2023, Volume and Issue: 36(9), P. 1444 - 1450

Published: Sept. 7, 2023

The use of quantum mechanics (QM) has long been the norm to study covalent-binding phenomena in chemistry and biochemistry. pharmaceutical industry leverages QM models explicitly covalent drug discovery implicitly characterize short-range interactions noncovalent binding. Predictive toxicology resisted widespread adoption QM, including industry, despite its obvious relevance metabolic processes upstream adverse outcome pathways advances both methods computational resources, which support fit-for-purpose applications reasonable timeframes. Here, we make case for embracing as an indispensable part a toxicologist's toolkit. We argue that provides necessary orthogonality alert-based expert systems traditional QSARs, consistent with calls animal-free integrated testing strategies safety assessments commercial chemicals. outline existing roadblocks this transition, need train model developers shift toward service-based toxicity utilize high-performance computing clusters. Lastly, describe recent examples successful implementations hazard propose how silico can be further advanced by integrating artificial intelligence.

Language: Английский

Citations

12
The rise of automated curiosity-driven discoveries in chemistry DOI Creative Commons
Latimah Bustillo, Teodoro Laino, Tiago Rodrigues

et al.

Chemical Science, Journal Year: 2023, Volume and Issue: 14(38), P. 10378 - 10384

Published: Jan. 1, 2023

The quest for generating novel chemistry knowledge is critical in scientific advancement, and machine learning (ML) has emerged as an asset this pursuit.

Language: Английский

Citations

12
Design Principles for Balancing Lipophilicity and Permeability in beyond Rule of 5 Space DOI
Henrik Möbitz

ChemMedChem, Journal Year: 2023, Volume and Issue: 19(5)

Published: Nov. 21, 2023

An ab initio conformational analysis of oral beyond Rule 5 (bRo5) drugs was complemented with measured permeability and logP(octanol) to derive design principles conferring bioavailability. 3D polar surface area (PSA) thresholds for bRo5 coincided those reported Ro5 space. The majority exceeded the logP threshold 5, reflecting a bias permeability. Above 500 Da molecular weight (MW), highly permeable Novartis compounds occupy narrow polarity range (topological or TPSA/MW) 0.1-0.3 Å

Language: Английский

Citations

11